Intracranial aneurysms are abnormal dilations of brain arteries that may rupture in some patients causing stroke and possibly death. Some treatment modalities are available, but they carry risks to patient health and hence best performed only on aneurysms that are at a high risk of growth or rupture. However, our understanding of factors associated with growth and rupture risk remains poor. Size is considered one factor, but widely acknowledged as a poor indicator. We hypothesize that aneurysm shape is a risk factor for growth and rupture risk on the basis of reported preliminary studies. The proposed project is the testing of this hypothesis. Our overall goal is to enhance understanding of the phenomenon of aneurysm rupture and to develop diagnostic analysis tools to facilitate clinical management of patients with intracranial aneurysms. There are three specific aims. Aim #1 is to develop the means to quantify the three-dimensional size and shape features of these lesions using techniques in computational geometry, computational hemodynamics and finite element tension analysis using computed tomography angiography (CTA) data. Aim #2 involves the determination of these indices in a study population of 75 ruptured and 75 unruptured consecutive aneurysms. The indices will then be compared between the two groups to identify the ones that differentiate ruptured from unruptured aneurysms. Aim #3 involves a five year longitudinal study of unruptured cerebral aneurysms where the indices chosen from specific aim #2 results will be statistically tested for predictability of growth during follow-up in the cohorts. The proposed project would be the first known assessment of three-dimensional geometry of brain aneurysms that integrates geometric, hemodynamic and solid mechanical methods. The computational tools to be developed can be of benefit beyond the scope of this project toward assessment of intracranial aneurysm treatment modalities and the mechanics of other similar disease processes. [unreadable] [unreadable] [unreadable]